about the constraints changed with variables

Awaiting user input

Hu Zhenghao

• July 17, 2022 03:03

Eexcuse me, in my problem, I have several constraints changed with variables. For example I need to achieve the model has following logic:

if p[0]>=4:

   model.addConstr(p[1]<=1.5)

else:

   model.addQConstr((p[1]-2.5)**2+(p[0]-4)**2<=16)

the logic of the problem is like this where p is my variable. How can I code this nonlinear constraint.

Thanks

 

1

• 

  Jonasz Staszek

  Community Moderator

  • July 18, 2022 06:19

  

  

  

  You can do it via indicator constraints. Perhaps this post will give you an idea of how to go about this.

  Additionally, I would suggest that you have a look at this entry concerning the creation of quadratic constraints, and make sure that your Python syntax is correct.

  Best regards
  Jonasz 

  0
• 

  Hu Zhenghao

  • July 18, 2022 12:07

  

  

  

  Thanks for your suggestion, I tried to use a binary variable and use indicator constraints, but the constraint is quadratic. So I added a new variable to define the quadratic equation, it works. I really appreciate your help!  

  0
• 

  Hu Zhenghao

  • July 18, 2022 12:45

  

  

  

  Actually, my objective is also changed with variables, like :

  if p[9,0]>=4:

     objective=max p[9,1]

  else:

     objective=max p[9,0]

  the expression is simplified.The objective function is segmented and nonlinear. Here the indicator constraints don't work, I also wonder how could I settle this.

  0
• 

  Jonasz Staszek

  Community Moderator

  • July 18, 2022 13:45

  

  

  

  I am afraid you can't change the objective this way. But perhaps you can find a way to reformulate your objective in such a way that the change is not required?

  Best regards,
  Jonasz

  0

Please sign in to leave a comment.